A Spam Filtering Method Based on Multi-Modal Fusion
Source abstract
Abstract
In recent years, the single-modal spam filtering systems have had a high detection rate for image spamming or text spamming. To avoid detection based on the single-modal spam filtering systems, spammers inject junk information into the multi-modality part of an email and combine them to reduce the recognition rate of the single-modal spam filtering systems, thereby implementing the purpose of evading detection. In view of this situation, a new model called multi-modal architecture based on model fusion (MMA-MF) is proposed, which use a multi-modal fusion method to ensure it could effectively filter spam whether it is hidden in the text or in the image. The model fuses a Convolutional Neural Network (CNN) model and a Long Short-Term Memory (LSTM) model to filter spam. Using the LSTM model and the CNN model to process the text and image parts of an email separately to obtain two classification probability values, then the two classification probability values are incorporated into a fusion model to identify whether the email is spam or not. For the hyperparameters of the MMA-MF model, we use a grid search optimization method to get the most suitable hyperparameters for it, and employ a k-fold cross-validation method to evaluate the performance of this model. Our experimental results show that this model is superior to the traditional spam filtering systems and can achieve accuracies in the range of 92.64–98.48%.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Spam and Phishing Detection · Network Security and Intrusion Detection · Internet Traffic Analysis and Secure E-voting
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
The School of Information and Software Enginerring, University of Electronic Science and Technology of China, Chengdu 610054, China
References (32)
Show source-provided references
1. Seth, S., and Biswas, S. (2017, January 4–7). Multimodal Spam Classification Using Deep Learning Techniques. Proceedings of the 2017 13th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), Jaipur, India.10.1109/SITIS.2017.91 ↗
2. Bettencourt, J. (2018, February 15). Kaspersky Lab Spam and Phishing Report: FIFA 2018 and Bitcoin among 2017’s Most Luring Topics. Available online: https://usa.kaspersky.com/about/press-releases/2018_fifa-2018-and-bitcoin-among-2017-most-luring-topics.
3. Carreras, X., and Marquez, L. (arXiv, 2001). Boosting trees for anti-spam email filtering, arXiv.
4. Androutsopoulos, I., Paliouras, G., and Michelakis, E. (2014). Learning to Filter Unsolicited Commercial E-Mail, National Center for Scientific Research. DEMOKRITOS.
5. A Bayesian approach to filtering junk e-mail
6. Anayat, S., Ali, A., and Ahmad, H.F. (2004, January 24–26). Using a probable weight based Bayesian approach for spam filtering. Proceedings of the 8th International Multitopic Conference 2004, Lahore, Pakistan.
7. Kim, H.J., Shrestha, J., Kim, H.N., and Jo, G.S. (2006). User action based adaptive learning with weighted bayesian classification for filtering spam mail. Australasian Joint Conference on Artificial Intelligence, Springer.10.1007/11941439_83 ↗
8. Yang, Z., Nie, X., Xu, W., and Guo, J. (2006, January 16–18). An approach to spam detection by naive Bayes ensemble based on decision induction. Proceedings of the Sixth International Conference on Intelligent Systems Design and Applications (ISDA’06), Jinan, China.10.1109/ISDA.2006.253725 ↗
9. Androutsopoulos, I., Paliouras, G., Karkaletsis, V., Sakkis, G., Spyropoulos, C.D., and Stamatopoulos, P. (arXiv, 2000). Learning to filter spam e-mail: A comparison of a naive bayesian and a memory-based approach, arXiv.
10. Optimizing semantic LSTM for spam detection
11. Abi-Haidar, A., and Rocha, L.M. (2008). Adaptive spam detection inspired by a cross-regulation model of immune dynamics: A study of concept drift. International Conference on Artificial Immune Systems, Springer.10.1007/978-3-540-85072-4_4 ↗
12. Shang, E.X., and Zhang, H.G. (2016, January 10–13). Image spam classification based on convolutional neural network. Proceedings of the 2016 International Conference on Machine Learning and Cybernetics (ICMLC), Jeju, South Korea.10.1109/ICMLC.2016.7860934 ↗
13. Wang, Z., Josephson, W.K., Lv, Q., Charikar, M., and Li, K. (2007). Filtering Image Spam with Near-Duplicate Detection, CEAS.
14. Kumar, P., and Biswas, M. (2017, January 9–10). SVM with Gaussian kernel-based image spam detection on textual features. Proceedings of the 2017 3rd International Conference on Computational Intelligence & Communication Technology (CICT), Ghaziabad, India.10.1109/CIACT.2017.7977283 ↗
15. Xu, C., Chiew, K., Chen, Y., and Liu, J. (2011). Fusion of text and image features: A new approach to image spam filtering. Practical Applications of Intelligent Systems, Springer.10.1007/978-3-642-25658-5_15 ↗
16. Huamin, F., Xinghua, Y., Biao, L., and Chao, J. (2011, January 3–4). A spam filtering method based on multi-modal features fusion. Proceedings of the 2011 Seventh International Conference on Computational Intelligence and Security, Hainan, China.10.1109/CIS.2011.100 ↗
17. Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012). Imagenet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, Curran Associates, Inc.
18. Graves, A. (2012). Long short-term memory. Supervised Sequence Labelling with Recurrent Neural Networks, Springer. Studies in Computational Intelligence.10.1007/978-3-642-24797-2 ↗
19. Dropout: a simple way to prevent neural networks from overfitting
20. Improved grid-based algorithm for Bader charge allocation10.1002/jcc.20575 ↗
21. Three way k-fold cross-validation of resource selection functions10.1016/j.ecolmodel.2007.10.005 ↗
22. Approximations to the log-likelihood function in the nonlinear mixed-effects model10.1080/10618600.1995.10474663 ↗
23. Bottou, L. (2010). Large-scale machine learning with stochastic gradient descent. Proceedings of COMPSTAT’2010;, Springer.10.1007/978-3-7908-2604-3_16 ↗
24. Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude
25. Kingma, D.P., and Ba, J. (arXiv, 2014). Adam: A method for stochastic optimization, arXiv.
26. Zhou, C., Sun, C., Liu, Z., and Lau, F. (arXiv, 2015). A C-LSTM neural network for text classification, arXiv.
27. Ioffe, S., and Szegedy, C. (arXiv, 2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift, arXiv.
28. Bouvrie, J. (2006). Notes on Convolutional Neural Networks, Unpublished.
29. Klimt, B., and Yang, Y. (2004). The enron corpus: A new dataset for email classification research. European Conference on Machine Learning, Springer.10.1007/978-3-540-30115-8_22 ↗
30. Dredze, M., Gevaryahu, R., and Elias-Bachrach, A. (2007). Learning Fast Classifiers for Image Spam, CEAS.
31. Gao, J., Lanchantin, J., Soffa, M.L., and Qi, Y. (arXiv, 2018). Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers, arXiv.10.1109/SPW.2018.00016 ↗
32. Robust biometric recognition from palm depth images for gloved hands10.1109/THMS.2015.2453203 ↗